Free Traders and Drug Smugglers:
The Effects of Trade Openness on
States' Ability to Combat Drug Trafficking
Horace A. Bartilow
Kihong Eom
ABSTRACT
The theoretical literature presents conflicting expectations about the
effects of trade openness on the ability of states to interdict drug
trafficking. One view expects that trade openness will undermine
drug interdiction; a second argues the opposite; a third argues that
trade openness does not necessarily affect drug interdiction. This
article assesses empirically the effects of trade openness on drug
interdiction for countries in the Americas using a pooled time-series
cross-sectional statistical model. It finds that trade openness
decreases the interdiction capabilities of states in drug-consuming
countries while increasing those of states in drug-producing countries. Greater openness to trade does not have a consistently significant effect on the interdiction capabilities of states in drug transit
countries.
I believe that the biggest problems to our security in the twenty-first century
and to this whole modern form of governance will probably come not from
rogue states or people with competing views of the wodd in governments,
but from the enemies of the nation-state, from terrorists and drug runners,
and organized criminals who, I predict, will increasingly work together and
increasingly use the same things that are fueling our prosperity: open
borders, (open commerce] and the Internet. . . . And we have to find a way
to cooperate, to deal with enemies of the nation-states, if we expect
progressive governments to succeed.
—President Biii Ciinton, November 21, 1999
C
ontroversy still swirls around the effects of trade openness on jobs
and wages in the industrialized world. Like President Clinton, some
scholars and policymakers view trade openness as fueling the prosperity
of nation-states, while others blame trade openness for massive job losses
and wage suppression (Lawrence 1996; Ahmad 2004; Teeple 1995). President Clinton's comment also underscores another controversy that
involves how organized criminal organizations, such as daig cartels,
exploit the opportunities of economic openness and ultimately threaten
the security of nation-states. How openness to trade affects the ability of
states to respond to this threat is the issue this article addres.ses.
2009 University of Miami
I
I
118
LATIN AMERICAN POLITICS AND SOCIETY
51: 2
Does the expansion of trade openness weaken the ability of states
to combat drug trafficking? Sotne researchers argue that trade openness
undermines the ability of states to effectively execute the "war on
drugs." Others contend that trade openness can enhance the ability of
states to combat the drug trade. Still others argue that trade openriess
does not necessarily affect states' ability to interdict drug trafficking.
While the various issues that surround trade openness and the security
probletns that are associated with drug trafficking have captured the
attention of policymakers and scholars, competing views about the
effects of trade openness on the interdiction capabilities of states have
been subjected to almost no large-N empirical investigation. The contributions of this study therefore are twofold: it offers an alternative perspective on how openness to trade affects states' ability to combat the
drug trade; and it systematically investigates these cotiipeting perspectives in the context of a large-N statistical design.
Like most global industries, the structure and operation of the illicit
drug industry is defined in terms of an international division of labor or,
more specifically, a narcotic international division of labor (NIIDL)
(Wallerstein 1974, 1984). Given certain endowments of geography! or
wealth, some countries find their niche as the drug industry's primary
producers; others find their niche as transit centers; and other countries
serve as the industry's principal consumers. This study argues that
depending on a country's niche in the NIDL, trade openness will have
different effects on states' ability to combat the drug trade. Trade openness will enhance the drug interdiction capabilities of states in drug-producing countries while weakening the interdiction efforts of states! in
drug-consuming countries. In countries that serve as transit centers for
drug smugglers, trade openness will have a limited effect on the diiug
interdiction capabilities of these states. This study empirically tests the
theory in the regional context of the Western Hemisphere. It utilizes a
statistical model that analyzes the effect of the interaction between trade
openness and countries' niche in the NIDL on the level of cocaine interdiction by states in the Americas from 1984 to 2003.'
After briefly reviewing the debate over the effects of trade openness
on the drug interdiction capabilities of states, the central hypotheses
derived from these perspectives are provided. This study contributes! to
this debate by introducing an alternative perspective. The study pr-oceeds with a discussion of the operationalization of the explanatory and
dependent variables, along with the problems and limitations of doing
empirical research using drug interdiction data, and how the research
design used here addresses these problems. The statistical model that is
used to analyze the data is presented, along with the detailed empirical
findings. The concluding section evaluates the policy implications of this
re.search and discusses future research efforts in this area of study. '
BARTILOW AND EOM: DRUG TRAEEICKING
\19
THE TRADE LDBEKAUZATION—DRUG
INTERDICTION CONTROVERSY
The existing literature on the effects of trade openness on the drug interdiction capabilities of states posits three competing propositions: trade
openness undermines the drug interdiction capabilities of states; trade
openness enhances the drug interdiction capabilities of states; and trade
openness does not necessarily affect the drug interdiction capabilities of
states. These views comprise the various arguments that are presented
below. The literature gives no theoretical consideration to the differentiation among countries in terms of how they are linked in the NIDL and
that given these differences, the effect of trade openness on the drug
interdiction capabilities of states may vary across countries. There may
be a fourth perspective, however, which demonstrates the importance
of taking these differences into consideration.
Trade Openness Undermines Drug Interdiction
The first argument under this rubric is that trade openness increases the
sheer volume of legal cross-border trade and provides cover for drug
smuggling by providing ample opportunity to conceal illicit cargo in licit
trade. Consequently, the problem for law enforcement is to detect illicit
commerce from licit trade. As trade openness keeps expanding the
volume of legal trade, states' ability to detect and interdict drug trafficking is severely diminished (Lepgold 1998; Williatns 1994a, 1994b; Treverton 1999; Serrano 2002; Stares 1996; Flynn 1993; Andreas 1998).
The large volume of containerized traffic mitigates against intrusive
inspections by border and customs officials. According to Paul Stares, in
the late 1990s, the top ten seaports in the wodd processed 33.6 million
containers, as opposed to 11.4 million in the previous ten years. Seaports in Singapore and Hong Kong processed more than 6 million containers, while seaports in Los Angeles and Rotterdam processed more
than 2 and 3 million containers respectively (Stares 1996). Today, the
sheer volume of trade via containerized shipping networks has allowed
drug traffickers to conceal illicit cargo and has significantly lowered the
probability of being detected by law enforcement.
According to the second argument, trade openness has fostered the
integration of financial markets and has provided drug traffickers with
more opportunities to launder money and invest in other activities, legal
or illegal, further strengthening the drug industry while weakening the
efforts of law enforcement to monitor the flow of drug money into the
legitimate economy (Andreas 1998; Stares 1996; Wisotsky 1986). In
recent years, drug traffickers have developed more sophisticated techniques of laundering drug proceeds through the legal economy. Today,
120
LATIN AMERICAN POLITICS AND SOCIETY
51.2
drug money is laundered through the stock market, Internet banking
and Internet casino, the insurance and real estate industries, credit and
debit card schemes, the diamond and gold industries, currency markets,
the entertainment industry, and the hotel and rental car industries
(Financial Action Task Force on Money Laundering 2003).
!
A third argument is that in the short and medium term, trade openness has a displacement effect on a country's work force. People who
have lost jobs due to trade reforms are more likely to turn to the drug
industry for alternative means of employment, thereby strengthening the
drug industry with an ample supply of recaiits (Hurrell and Woods 1999;
Held and McGrew 2000; George 1992; Taylor 2002). Trade openfiess
reforms that remove preferential trade policies, like the Furopean
Union's Lomé Trade Accords of 1975, which were designed to benefit
the commodity exports of African, Caribbean, and Pacific counti-ies,
have significantly reduced the income of producers in developing countries; these producers could turn to the drug industry for alternative
income (Jam.aica Daily Gleaner 1996).
The fourth view from this perspective argues that by fosteringi the
expansion of trade and global transportation networks, trade openness
has also encouraged the cooperation and the formation of alliances
among criminal organizations across different countries. Such cooperation expands the power and reach of cartels to distant markets and polities and thereby strengthens their ability to evade detection by local|law
enforcement (Williams 2002; Clawson and Lee 1996; Williams 1994b;
Kerry 1998). It has been reported that criminal organizations work
together to coordinate their money-laundering activities by havingl different criminal organizations handle specific stages of the laundering
process. One organization stnictures the process through which financial transactions will be laundered, while another criminal group provides the "dirty" money to be cleaned. Moreover, a criminal organization that may have laundered money for another organization in lone
instance may reverse roles and provide its own "dirty" money to|that
organization for laundering on another occasion (Financial Action Task
Force on Money Laundering 2002). The arguments from this perspective
generate the following proposition:
'
H.I: Increasing levels of trade openness enhance the ability of drug
traffickers to avoid interdiction by states.
Trade Openness Enhances Drug Interdiction
The first argument from this perspective is that the financial gains from
trade will reduce the incentive to participate in the drug industry. Trade
openness can open new markets to domestic producers who would !oth-
BARTILOW AND EOM: DRUG TRAFFICKING
121
erwise resort to exporting illicit drugs (Martin 1993; Dillin 1989; Clawson and Lee 1996; Lee 1989),^
The second view argues that trade openness will increase revenues
to states via an influx of foreign direct investment. Increases in government revenues can be used to improve the overall drug enforcernent
capabilities of states. Increased revenues can also be used to purchase
modern radar and surveillance technology and to improve the training
and intelligence-gathering operations of drug enforcement personnel.
Moreover, expanded revenues would also enable governments to provide higher wages and better benefits to law enforcement personnel,
making them less vulnerable to bribes and the corruptive lure of drug
money (Mabry and Perl 1989; U,S, Department of Defense 1991; Anderson 1988).
The third view argues that extensive trade openness among states
can encourage greater cross-border drug enforcement coordination
between law enforcement agencies in different countries. According to
proponents of the North American Free Trade Agreement (NAFTA), one
of the benefits of that trade accord is the increased antinarcotic cooperation between the United States and Mexico, Similarly, European integration has made possible the institutionalization of police and judicial
cooperation across Europe in combating drug trafficking and other
transnational crimes. While trade openness facilitates the integration of
markets, it also strengthens cross-national cooperation in the gathering
and sharing of intelligence that is relevant in the interdiction and prosecution of drug smugglers (Anderson et al. 1995; McLaughlin 1992; den
Boer 2002; den Boer and Walker 1993), The arguments from this perspective generate the following proposition:
H,2: Increasing levels of trade openness will increase the ability of
states to interdict drug trafficking.
Trade Openness Does Not Necessarily
Affect Drug Interdiction
The first argument from this perspective is that the extent of trade expansion is grossly exaggerated, because world trade, as a percentage of gross
domestic product, is roughly the same today as it was before World War
I, There is no evidence to suggest that the ratio of illegal to legal trade is
increasing. In fact, given the degree to which trade flows have become
liberalized, the opportunities for transnational crime could be declining
relative to the total level of international economic exchange. Therefore
it is not clear what effect trade openness does have on transnational
crime, such as drug trafficking, and consequently on the ability of law
enforcement to interdict drug smuggling (Naylor 2002; Serrano 2002).
122
LATIN AMERICAN POLITICS AND SOCIETY
51: 2
The second perspective is that while the openness of financial markets is vital to world trade, there is no consistent evidence that a change
from a restricted exchange-control market to a liberalized market has
had more than a modest effect on the level of financial crimes like
money laundering. The openness of financial markets has had a negligible effect on the amount of money laundered, and has not necessarily weakened the efforts of law enforcement in combating illicit financial transfers from drug trafficking (Levi 2002; Naylor 2002).
'
The third view argues that the much-touted relationship between
trade openness and the rise in drug trafficking may actually be spuricius,
because the expansion of legitimate trade has been accompanied by| an
equivalent increase in the selective criminalization of trade via government prohibitions. Therefore, trade openness may not have any effect
on the ability of states to interdict drug trafficking (Serrano 2002). •
The fourth view argues that while trade openness has made modern
commerce more dependent on advanced communications and transportation technology than ever before, the impact of technology ion
illicit transfers today and on the ability of governments to interdict
smuggling may be no greater than those produced by the introduction
of the railroad and the automobile in the early decades of the twentieth
century (Naylor 2002; Serrano 2002). The arguments from this perspective generate the following proposition:
H.3: Increasing levels of trade openness will have no effect on 'the
ability of states to interdict drug trafficking.
THE EFFECT OF TRADE OPENNESS ON
DRUG INTERDICTION VARIES ACROSS COUNTRIES
The essential limitation of the existing literature is that it assumes dhat
the effects of trade openness on drug interdiction will be uniforni in
spite of the different niches that countries occupy in the NIDL. In otber
words, the literature predicts that trade openness will have either a positive, a negative, or no effect on the ability of states to combat drug trafficking. This study offers an alternative perspective, which posits dbat
the effects of trade openness on drug interdiction will vary depending
on a country's niche in the NIDL.
Trade Openness and Interdiction in
Drug-Producing Countries
Counterfactually, in the absence of trade openness, indigenous cartels
would displace legitimate firms from the production process. The
money-laundering activities of drug cartels can result in the cheapening
BARTILOW AND EOM: DRUG TRAFFICKING
123
of hard currency relative to the local currency. The narco-induced
appreciation of the exchange rate can reduce the competitiveness of the
products of legitimate firms in domestic and foreign markets, which will
ultimately reduce their market share and, in the process, shrink the tax
base that finances the state's drug enforcement policies. Moreover, in
the absence of trade openness, the growth of the drug industry does not
provide any backward or forward linkages to the legal economy and
does very litde to stimulate economic growth in the legitimate sectors of
the economy (Clawson and Lee 1996).
Increasing levels of trade openness, by contrast, is likely to encourage joint venture investments between local and transnational firrns and
thereby provide access to foreign markets for local products and boost
economic growth and domestic employment—all of which discourages
participation in the drug industry and adds revenue to the state's coffers
to finance its counternarcotics operations (Johnson 1970, 55).
By increasing the number of joint venture investments between
local and transnational firms, moreover, trade openness will efficiently
mobilize resources that would otherwise flow to the illicit economy
(Biersteker 1981, 2). Relative to domestic firms, transnational corporations have a greater capacity to mobilize domestic capital, labor, and the
technology needed for production. It is more likely that in the absence
of trade openness, more domestic resources would flow to the drug
industry and would not be used productively in the legitimate economy
(Vernon 1975, 20-21). Therefore, in mobilizing domestic resources for
legitimate production, trade openness would not only increase the
state's tax base but also reduce resources that would be available to the
drug industry. This discussion generates the following proposition:
H.4: Increasing levels of trade openness will increase tbe ability of
states in drug-producing countries to interdict drug trafficking.
Trade Openness and Interdiction in
Drug-Consuming Countries
The consumption of illicit drugs in many drug-producing countries in
Latin America is significantly less than in the United States. According to
Colombia's National Drug Council, in 1994, just 1.4 percent of the population, or 1.4 million people, had consumed illicit cocaine at some time
in their lives, while in the United States, drug consumption among lifetime users involved 31 percent of the population, or 77 million people
(Colombian National Drug Council 1994). According to the United
Nations World Drug Report 2006, since the 1980s, the United States
remains the largest single cocaine market, accounting for more than 40
percent of all cocaine users in the world.-'' The annual prevalence of
124
LATIN AMERICAN POLITICS AND SOCIETY
51: 2
cocaine use between ages 15 to 64 in the United States is 2.8 percent of
the population, or 6.38 million people. The annual prevalence of
cocaine use in South America (including Central America and the
Caribbean) for the same age range is only 0.7 percent of the population.
The largest opiate market in the Americas is also the United States, with
approximately 1.2 million heroin users (United Nations Office on Drugs
and Crime 2006).
I
The strong U.S. demand for illicit drugs, coupled with the nation's
relatively high levels of trade openness, will increase the sheer volume
of licit and illicit trade, which is likely to overwhelm the interdiction
capabilities of the state. During the 1980s and 1990s, it took five IJJ.S.
customs officers three hours to inspect one container (Stares 19^6).
Since the terrorist attacks of September 11, 2001, U.S. Customs and
Border Protection has significantly improved the speed and efficiency of
port and cross-border inspections with the purchase of modern, automated systems and sophisticated surveillance technology; but the same
two factors continue to make drug smuggling into the United States a
lucrative enterprise and provide smugglers with new and creative ways
of evading the most sophisticated detection technologies. Given the
large volume of trade that comes into the United States, Raymond W.
Kelly, former GBP commissioner, notes, "finding contraband hidden
among hundreds of millions of travelers and billions of dollars of trade
is literally like finding a needle in a haystack" (U.S. Gustoms and Border
Protection 2000a). Whenever U.S. Gustoms uncovers one method of
concealment by drug smugglers, it merely forces them to take alternative measures to evade detection (U.S. Gustoms and Border Protection
2000b). This discussion generates the following proposition:
I
H.5: Increasing levels of trade openness will reduce tbe ability ofstates
in drug-consuming countries to interdict drug trafficking.
Trade Openness and Interdiction in
Drug Transit Countries
i
I
The geography of the Western Hemisphere makes Central American and
Caribbean countries convenient transshipment sites for South American
drug producers that target markets in the United States and Canada.
According to U.S. policymakers, in the 1990s, the primary security threat
to the United States was the transshipment of large amounts of cocaine
through the Caribbean Basin (Ped 1993-94; Tokatlian 1994). How
would trade openness affect the interdiction capabilities of states in transit countries? It is argued that while Caribbean Basin countries have
gained from various trade openness arrangements, such as the
Caribbean Basin Initiative (GBI) and the General System of Preferences
BARTILOW AND EOM: DRUG TRAFFICKING
125
(GSP), their economies have been significantly disadvantaged by free
trade arrangements made outside the region.
NAFTA, for example, has diverted Caribbean Basin trade from the
North American market. Trade economists predicted that under NAFTA,
the Caribbean Basin would experience a significant displacement of its
textile and apparel exports to North America (Hufbauer and Schott
1992). The preferential concessions under the CBI and the GSP to promote export diversification among Central American and Caribbean
Basin countries has been largely undermined by Mexico's preferential
access to the U.S. and Canadian markets under NAFTA (Hutchinson and
Schmacher 1996). As a direct result of trade and investment diversion,
the Caribbean Basin has suffered the closing of more than 150 apparel
plants and the loss of more than 123,000 jobs. With the inception of
NAFTA, Mexican textile exports to the U.S. and Canadian markets have
grown at three times the rate of those from the Caribbean Basin. In
Jamaica, garment exports fell by 7 percent, resulting in the loss of more
than 7,000 jobs. Guyana, Belize, and St. Lucia have recorded even larger
job losses (Rother 1997). The Wodd Trade Organization ruling against
the Lomé Convention, which was established to give Caribbean Basin
banana and sugar exporters preferential access to the Furopean market,
also reduced the competitiveness and export earnings of the region's
producers (Dodwell and James 1993).
Thus, while the Caribbean Basin countries have gained from free
trade arrangements like the CBI and the GSP, these gains were offset by
later losses that were attributed to free trade policies emanating from
outside the region. On balance, trade openness may not generate the
level of revenue that is needed to finance effectively the counternarcotics operations of the state. It is therefore expected that trade openness will not have a significant impact on the drug interdiction capabilities of states in the transit countries of Central America and the
Caribbean.'* This discussion generates the following proposition:
H.6: Increasing levels of trade openness will have no effect on the ability of states in drug transit countries to interdict drug trafficking.
MEASUMNG THE EFFECTS OF TRADE OPENNESS
ON DRUG INTERDICTION
Hypotheses 1 through 6 can be tested by estimating drug interdiction
utilizing a pooled time-series cross-sectional analysis of the data for all
member countries of the Organization of American States, spanning the
period 1984—2003.^ The central explanatory and dependent variables of
the research are operationalized; then the limitations of empirical
research using narcotic interdiction data are discussed. The research
126
LATIN AMERICAN POLITICS AND SOCIEIT
51:2
design in this study, which includes a battery of theoretically and empirically motivated proxy variables, attempts to solve these limitations.
Primary and Dependent Variables
The primary explanatory variable is trade openness and the country's
niche in the NIDL. Following convention, this study adopts a measure
of trade openness that is featured in many political and economic analyses of trade and financial flows in the global economy (Li and Reuveny
2003; Dreher 2006). The variable is taken from the World Bank's World
Development Indicators 2006 and is computed as the yearly aggregate
of a countiy's total imports and total exports (in millions of U,S, dollars)
divided by its GDP in constant 1995 U,S, dollars. This variable measures
the extent to which a country's economy is open if a large share of its
GDP comes from trade (World Bank 2006).
The trade openness variable interacts with a trichotomous variable,
which represents differences in a country's niche in the NIDL. DPC is
coded 1 if a country is a primary drug-producing country and 0 otherwise, DTC is coded 1 if a country is a primary drug transit country and
0 otherwise. DCC is coded 1 if a countiy is a primary drug-consuming
country and 0 otherwise. These codes were adopted from various issues
of the CIA's World Pact BookJ'
'
The dependent variable is the level of drug interdiction, operationalized as the yearly interdiction of cocaine measured in kilogranis.
The drug interdiction data were collected from the OAS Inter-Ameridan
Drug Abuse Control Commission (CICAD n,d.).
Control Variables
Number of Drug Enforcement Personnel Researchers in sociology and
criminal justice have provided theoretical and empirical reasons for considering differences in the size of law and drug enforcement agencies in
combating the daig trade (Tulder 1992), In following these researchers,
it is considered that differences in the size of government agencies
responsible for drug enforcement would affect the level of drug interdiction. Specifically, it is expected that countries with more drug enforcement personnel are more likely to have higher levels of interdictions than
in countries where the number of drug enforcement personnel is small,^
The drug enforcement personnel data were adopted from various issues
of The Military Balance (Institute for Strategic, Studies 1984-2006).
U.S. Aid and Bilateral Drug Maritime Interdiction Agreements with
the United States. An important factor that affects the interdiction capabilities of governments in the Western Hemisphere is the provision of
aid to countries that U,S, policymakers deem to be cooperating with the
BARTILOW AND EOM: DRUG TRAFFICKING
127
United States in combating drug trafficking. U.S. aid to support counternarcotics operations may have a positive impact on the amount of
drugs that recipient countries actually interdict. This study therefore
controls for the level of U.S. aid given to Western Hemisphere countries
to support drug interdiction operations. Data for this variable were collected from the U.S. Agency for International Development, U.S. Overseas Loans and Grants.*^
We also control for whether countries in the region are parties to a
maritime drug interdiction agreement with the United States, because
such agreements may enhance the interdiction capabilities of governments in the Western Hemisphere. Bilateral maritime agreements with
the United States allow for joint sea and helicopter searches of vessels
involved in transporting illegal drugs witbin the territorial waters of
respective governments. These agreements also permit land and sea
patrols by the U.S. Coast Guard and Navy, as well as maritime searches
and seizures and drug arrests by U.S. law enforcement authorities within
the sovereign boundaries of countries in the region (DEA Strategic
South American/Caribbean Unit 2001). This indicator is a dichotomous
variable that measures whether or not countries in our sample, in any
given year, are signatories to a marititne interdiction agreement with the
United States. This variable is coded 1 if a country is party to this agreement and 0 otherwise. The variable was collected from the U.S. Department of State and various issues of the International Narcotics Control
Strategy Report (U.S. State Department 1988, 1992).
Money Laundering and Government Corruption. Interviews with
DEA financial investigative agents suggest that the extent to which a
country's banking and financial sectors become instruments for smugglers to conceal, move, and launder drug proceeds is an important
factor that undermines states' drug interdiction capabilities (Walters
2001). Money laundering represents the corruption of a country's
financial system. While the various service providers of a country's
financial sector may not know the criminal nature of the proceeds they
help to move or conceal, others launder proceeds with full knowledge
of the fund's illegal origin (Financial Action Task Force on Money
Laundering 2002). Because drug traffickers value the convenience of
easily laundering drug proceeds while smuggling drugs through various countries, it is argued that states whose banking and financial sectors fail to implement internationally recognized anti—money-laundering measures as established by the Financial Action Task Force are less
likely to interdict drug trafficking. Since money laundering may
weaken the drug interdiction capabilities of states, we control for this
phenomenon.
Government corruption can grease the wheels of drug trafficking by
improving the efficiency of smuggling, enabling traffickers to smuggle
128
IJVTIN AMERICAN POLITICS AND .SOCIETY
51:2
drugs in and out of countries with itnpunity (Leff 1964; Lui 1985; Kaufmann and Wei 1999; Galtung 2000). Drug traffickers are therefore more
likely to move their drugs through countries where government corruption is embedded in the normal operations of the state and its agents,
as opposed to countries where government agents and institutions are
relatively clean. Since corruption undermines the integrity of the state
and its officials, high levels of corruption are expected to have a negative effect on a government's ability to interdict drugs.
;
The data that measure whether a country is cooperating to implement international anti-money-laundering measures were collected from
various issues of the FATF Report on Money Laundering Typologies
(2001-3). This variable is coded 1 if a country in any given year is listed
by the FATF as failing to implement international anti-money-laundering
measures or 0 otherwise.' The corruption variable for countries in o,ur
sample was collected from the Transparency International Perception
Comiption Index. Scores on the index range from 0 (which indicates a
country where business transactions are entirely penetrated by government corruption) to 10 (which indicate a perfectly clean government).
i
BEHAVIORAL FACTORS
,
To test the differential effects of trade openness on drug interdiction,] it
is important to consider the factors that shape the behavior of drug traffickers, since the behavior of traffickers affects states' ability to interdict
drug trafficking. But because drug trafficking is an illicit enterprise, tlie
behavior of drug traffickers is unobservable, and therefore cannot be
measured directly (OECD. Statistics Directorate et al. 2002). Therefore,
a significant limitation of empirical research on narcotics interdiction
research is that it cannot be observed how traffickers adjust their behavior in response to the interdiction measures of drug enforcement officials (Crawford et al. 1988). If drug enforcement forces in Costa Rica, for
example, rigorously patrol the country's territorial waters and air space,
then traffickers seeking to penetrate the U.S. market could simply moye
their smuggling operations through Haiti, where drug enforcement surveillance may be less rigorous (Adams 2002; Green 2002).
In essence, empirical research on drug interdiction must address the
problem of spatial correlation. This simply means that the interdiction
of drugs across countries is related. This study therefore constructed a
variable that is intended to capture the effects of spatial correlation on
drug interdiction throughout the Americas. This variable is defined as
the Drug Enforcement Effort of Neighboring Countries in the region (for
the formula, see the appendix).
Another limiting factor is that because drug trafficking is unobservable, we cannot observe what portion of the total amount of drugs
BARTILOW AND EOM: DRUG TRAEFICKING
129
being smuggled is actually interdicted. High levels of interdiction may
reflect the reality that larger rather than smaller amounts of drugs are
being smuggled into countries (Bagley 1989; Stares 1996). Research on
narcotics interdiction is therefore limited by imperfect data because
interdiction itself is imperfect.'°
In addressing this limitation, economists in recent years have developed methodological models that utilize proxy variables that indirectly
measure the size of the underground economy (OEGD Statistics Directorate et al. 2002; Tanzi 1982a; Pozo 1996; Frey and Pommerehne 1982).
The design of the prsent study draws on the basic logic that has
informed the methodological tradition of this research. This study compensates for data imperfection by controlling for a package of proxy
variables that attempt to capture the incentives that shape the underlying structure of drug trafficking (Molefsky 1982; Feige 1996). However,
this design goes beyond the methodological design of previous research
in this tradition by generating proxy variables that have emerged from
extensive interviews with U.S. DFA special agents and with coast guard
and drug enforcement agents in Ganada, Jamaica, Trinidad and Tobago,
and Golombia. With the permission of officials at the Federal Gorrectional Gomplex in Goleman, Florida, interviews were also conducted
with former drug traffickers who were willing to share their insights into
the structure and complexities of drug smuggling.
The proxy variables those respondents identified include geography, per capita national income, the level of civil liberties and political
rights, and the level of political competitiveness in a country's polity.
Geographic Coastline and Distance
Former drug traffickers and the drug enforcement officials who were
interviewed consistently noted that drug trafficking is as much about
covering distance as it is about generating profit. Given the importance
of geography for drug cartels, the geographic location and physical
characteristics of countries make them more or less conducive to drug
trafficking. Interview subjects noted that the length of a country's coastline is an important aspect of the incentive structure that shapes the
level of smuggling. Drug trafficking through countries with long coastlines reduces the chances of detection by the coast guard compared to
countries with shorter coastlines, which increase the likelihood of detection (Green 2002; Adams 2002; Lewin 2006; Edwards 2006).
All things being equal, interview subjects also noted that longer
coastlines reduce the chances for interdiction. They said that the geographical distance between drug-producing and transit countries and
destination markets in drug-consuming countries in the Americas,
specifically the United States and Ganada, is also an important aspect
130
lj\TIN AMERICAN POLITICS AND SOCIETY
51:2
of the incentive structure that shapes the underlying level of drug ti"afficking. Traffickers are more likely to move their drugs through countries that are closer to destination markets in North America than
through countries that are farther away (Green 2002; Adams 2002;
Hinds 2006; Batchelor 2006). Relative to countries that are closei] to
destination markets, it is expected that countries that are farther away
will be less exposed to trafficking. Therefore, interviewees noted that
greater distance from destination markets reduces the chances ¡for
interdiction.
The coastline variable for the countries in our sample was collect:ecl
from the CIA World Fact Book 2006, in which a country's total coastline
is measured in kilometers. The distance variable measures the average
geographical distance between the capitals of drug-producing and transit countries, and the capitals and land contiguity of drug-consuming
countries, reported in miles. This variable was taken from Direct-Line
Distances, U.S. Edition (Fitzpatrick and Modlin 1986).
'
National Income Per Capita
Former traffickers noted that the level of wealth among the countries
through which they smuggled was a critical aspect of the incentive
stRicture of drug trafficking (Green 2002; Adams 2002). Since a portion
of the drugs that are manufactured in drug-producing countries and
smuggled through transit countries is distributed to local users before
reaching the primary customers in consuming countries, wealthier coiintries in the Americas are likely to spend more on illicit drugs and toibe
exposed to higher levels of trafficking than poor countries in the region.
Therefore, the interview subjects noted that wealthier countries in the
region that attract more trafficking increase the chances for interdiction.
The study measures a country's per capita national income in terms of
the per capita income purchasing power parity at current international
prices. Data for this variable were collected from the World Bank's
World Development Indicators, 2006.
\
Civil Liberties and Political Rights
The spread of economic openness has also given rise to the spreacj of
democratization and individual political freedom (Gill 1996; Monshipouri 1995; Mittleman 1996; Rueschemeyer and Evans 1985; Drake
1998). The Canadian and U.S. drug enforcement agents interviewed said
that the level of individual political freedom, defined in terms of the
level of political and civil liberties of citizens, is an important aspect of
the incentive structure that shapes the level of drug trafficking (Helson
2004; Ogden 2006).
BARTILOW AND EOM: DRUG TRAFFICKING
'
131
These agents noted that in countries where political and civil liberties
are relatively high, citizens are protected from arbitrary arrest and
searches by the state (Craig 1980). While these rights protect the civil liberties of citizens, they also constrain drug enforcement officers' ability to
combat the drug trade (Wisotsky 1986, 117-39; Ogden 2006; Cabana
2006), Drug trafflckers are therefore more likely to smuggle drugs through
countries where political rights and civil liberties are relatively high, as
opposed to countries where individual political freedoms are low and the
state is not constrained by the civil liberties of its citizens. Consequently,
Canadian and U,S. daig enforcement personnel noted that high levels of
political freedom and civil liberties could shield dmg trafflckers from arbitrary state power and thereby reduce the chances for interdiction."
The measure of individual political freedoms and civil liberties for
the countries in our sample was collected from the Freedom House
compilation Ereedom in the World Country Ratings, 1972-2007 (Freedom House 2007), The scores are rescaled from 1 to 7, where 1 is the
lowest score for political rights and civil liberties and 7 is the highest.
Level of Political Competitiveness
Drug enforcement officials from Colombia, Jamaica, and Trinidad and
Tobago who were interviewed also reported that the political competiveness (POLCOMP) of a country's polity is an important aspect of
the incentive structure that shapes the level of drug trafflcking. Drug
traffickers are more likely to smuggle drugs through countries where the
political system enables nonelites to influence political elites in regular
ways. The more open countries' national polities become to nonelite
participation, the easier it is for drug traffickers to gain access to political leaders and potentially to influence the decisions of political elites
(Garcia 2002; Hinds 2006; Paul 2006; Craig 2006; Watkis 2006),
In many Western Hemisphere countdes, drug trafflckers have participated in the competitive political process by making significant flnancial contributions to political parties in an attempt to co-opt political
candidates (U.S, State Department 1992). In many of these countries,
drug-trafficking organizations maintain loose affiliations with the country's politicians in order to secure protection from local law enforcement
(Rosenberg 2000; U.S, State Department 1988).'^ Therefore, the drug
enforcement offlcials interviewed noted that countries with high levels
of political competition will be exposed to high levels of drug trafficking, and therefore the chances for drug interdiction will increase.
The level of political competiveness in a country in any given year
is collected from the POLITY IV dataset (Jaggers and Gurr 1993; Polity
IV Database). The variable POLCOMP is an index that ranges from 1 to
10. A score of 1 is attributed to polities that are closed, A score of 10 is
132
LATIN AMERICAN POLITICS AND SOCI BIT
51:2
I
attributed to polities are relatively open. Political groups and organizations in these polities regularly compete for political influence with little
use of coercion (Marshall andjaggers 2002).
THE STATISTICAL MODEL AND EMPIRICAL
RESULTS
'
i
The theoretical model as discussed can be written as a regression equation. To evaluate the different intercepts and slope coefficients, indicator variables are inserted and interact with the trade openness variable.
The indicator variables are drug-producing countries (DPC), drug transit countries (DTC), and drug-consuming countries (DCC), where DPC
is used as a base category (Berry and Feldman 1985). The full equation
can be written as follows:
Daiglnterdiction.^ = ßg + ß,DTC., +
ß^NeighborEnforcementEffortjj + ß^Coast., +
ßioPPPi, + ßuRights, + ßj^Polcomp., +
ßijDrugEnforcementPersonnel.j + ß
tion.^ + ßjgBilaterallnterdiction.j + EJ,
The estimation method is ordinary least squares (OLS) with panel
corrected standard errors (PCSE). The assumptions on the error term are
£(e^.,) = G], £(£., e^.,) = a^., and Eiz., e^.,.) = 0 (Beck 2001; Beck Jnd
Katz 1995).
The six hypotheses that predicted the effects of trade openness ion
drug interdiction were empirically tested. Table 1 shows that the interaction terms for Trade Openness-Drug-Producing Countries, and Trade
Openness-Drug-Consuming Countries are statistically significant with
the signs of the coefficients in the predicted direction. Consistent with
the predictions of the alternative hypotheses, increasing levels of trade
openness in the Western Hemisphere have a positive effect on the drug
interdiction capabilities of states in drug-producing countries but a negative effect on the interdiction efforts of states in drug-consuming countries, and no statistically significant effect on the interdiction capabilities
of states in transit countries.'^ These results suggest that in the Western
Hemisphere, the effect of trade openness on drug interdiction varies
across countries depending on their niche in the NIDL.
BARTILOW AND EOM: DRUG TRAFFICKING
133
Table 1. Trade Openness and Drug Interdiction in the Americas,
by County's Niche in INDL
Drug transit countries
Drug-consuming countries
Trade openness-drug-producing countries
-0.328
(0.470)
2.134"*
(0.610)
0.091"*
(0.017)
Trade openness-drug transit countries
-0.000
(0.002)
Trade openness-drug-consuming countries
-0.083***
(0.016)
Drug enforcement personnel
0.000***
(0.000)
Total coastline (km)
0.001***
(0.000)
Distance from drug-consuming countries
-0.040***
(0.007)
Per capita purchasing power parity
0.133***
(0.040)
Civil liberties and political rights
-0.743***
(0.083)
Government comiption
0.175***
(0.068)
Money laundering
-0.414**
(0.195)
0.328***
(0.063)
0.008
(0.048)
Political competiveness
Neighboring country's enforcement effort
Bilateral maritime interdiction agreement
0.008
(0.027)
USAID
0.000
(0.001)
Constant (drug-producing countries)
7.290***
(0.797)
Obsei-vations
400
0.52
Panel corrected standard errors in parentheses.
*Significant at 10%, **significant at 5%, ***significant at 1%
134
1J\TIN AMERICAN POLITICS AND SOCIE'n'
51: 2
In terms of our control variables, it was argued that U.S. political
influence, specifically its use of aid to encourage foreign government
cooperation in combating drug trafficking, is expected to have a positive effect on drug interdiction throughout the Western Hemisphere.
However, the variable USAID is not statistically significant.''' As
expected, the variable Drug Enforcement Personnel, which measures
the size of countries' drug enforcement agencies, has a positive and statistically significant effect on drug interdiction. Also as expected, the
prevalence of money laundering has a negative and statistically significant effect on drug interdiction. The prevalence of government integrity
(low government corruption) has a positive and statistically significant
effect on drug interdiction. It was also argued that the interdiction of
drugs across countries is related. The variable Drug Enforcement Effort
of Neighboring Countries, which controls for the effects of spatial correlation, is not statistically significant. Gountries' bilateral maritime interdiction agreements with the United States have no statistically significant
effect on drug interdiction.
How did the proxy variables perform in capturing the incentives
that structure the underlying level of drug trafficking? In terms of geo'graphy, we argued that longer coastlines reduce the chances for interdiction and that greater distance from destination markets also reduces
the chances for interdiction. The effect of a country's coastline on drug
interdiction is statistically signiflcant, but the sign of the coefficient is not
consistent with the predicted direction. However, the effect of a country's distance from destination markets in North America on drug interdiction is statistically significant, with the sign of the coefficient in the
predicted direction. We also argued that richer countries are more likqly
to attract more trafficking and therefore increase the chances for interdiction. As predicted, the effect of a country's per capita purchasing
power parity on drug interdiction is positive and statistically significant.
We argued that high levels of political freedom and civil liberties
will shield drug traffickers from arbitrary state power and thereby
reduce the chances for interdiction. As predicted, the effect of civil liberties and political rights on drug interdiction is negative and statistically
significant. It was also argued that countries with high levels of political
competition would be exposed to more trafficking, which would
increase the chances for interdiction. Gonsistent with the prediction, the
effect of a country's political competitiveness on drug interdiction • is
positive and statistically significant.
'
Some readers may object to our adoption of the GlA's coding, which
classifies countries in our dataset as drug producing (or drug source
countrie.s), drug transit, and drug consuming; they may think that our
findings are an artifact of these classifications. In testing the validity 'of
this concern, we relaxed our theoretical argument and ran separate
BARTILOW AND EOM: DRUG TRAFFICKING
135
Table 2. Trade Openness and Drug Interdiction in the Americas,
by Regional Grouping
Explanatory
Variables
Andean Non-Andean Central
South
South
America and North
Caribbean America
America
America
Trade openness
0.110***
(0.041)
0.002***
(0.001)
0.001
(0.001)
-^.071**
(0.029)
Drug enforcement personnel
0.003*
(0.002)
-0.000
(0.000)
0.000***
(0.000)
0.000
(0.000)
Total coastline (km)
0.916**
(0.435)
0.032***
(0.006)
0.045***
(0.007)
-0.002***
(0.001)
Distance from drug-consuming 0.638
countries
(0.417)
-0.141***
(0.020)
0.071***
(0.019)
0.000
(0.000)
Per capita purchasing power
parity
-0.565*
(0.341)
0.418***
(0.050)
0.030
(0.039)
-0.128*
(0.075)
Civil liberties and political
rights
-1.277***
(0.298)
-0.257
(0.157)
-0.457***
(0.049)
0.000
(0.000)
Government corruption
0.411**
(0.202)
0.018
(0.117)
0.051
(0.063)
1.543***
(0.438)
Money laundering
0.000
(0.000)
0.000
(0.000)
0.308
(0.188)
0.000
(0.000)
Political competitiveness
0.370*
(0.223)
-0.049
(0.070)
0.337***
(0.066)
0.000
(0.000)
Neighboring country drug
enforcement
-0.324**
(0.163)
-0.525***
(0.155)
0.279***
(0.03.3)
0.361**
(0.141)
Bilateral maritime interdiction
0.197
(0.436)
0.849**
(0.377)
-0.140
(0.145)
0.077*
(0.040)
USAID
-0.000
(0.001)
0.031***
(0.012)
-0.007**
(0.003)
0.000
(0.000)
-34.872
(27.266)
12.406***
(1.520)
1.148*
(0.678)
21.521***
(5.042)
Constant
Observations
51
0.81
94
0.83
227
0.52
28
0.96
Andean countries include Peai, Bolivia, Ecuador, Colombia, and Chile. Non-Andean
countries: Argentina, Brazil, Uruguay, Paraguay, and Venezuela. North American
countries: the United States, Canada.
Panel corrected standard errors in parentheses.
*Significant at 10%, **significant at 5%, ***signincant at 1%
136
LATIN AMERICAN POLITICS AND SOCIE'IT
51:2
time-series cross-section regressions without the interaction terms for a
country's niche in the NIDL, which is based on the GIA's classification.
These regressions produced estimates for the effects of trade openness
on drug interdiction for the countries of South America (Andean and
non-Andean), Gentral America and the Garibbean, and North America.
Even after respecifying the model in this way, the results, as reported in
table 2, are robust and substantively consistent with those reported in
table 1, which uses the GIA classification.'^
|
Trade openness has a positive and statistically significant effect on
the interdiction capabilities of states in South America (both Andean and
non-Andean countries). Among these countries, Golombia, Bolivia, and
Peru are considered to be the world's primary source countries for
cocaine production. Trade openness has no statistically significant effect
on the interdiction capabilities of states in Gentral America and the
Garibbean, countries widely regarded as drug transit centers. Traele
openness has a negative and statistically signiflcant effect on the interdiction capabilities of states in North America. Among these countries,
the United States is widely recognized as the world's largest consumer
of cocaine.
I
I
CONCLUSIONS
;
The theoretical literature presents conflicting expectations about the
effects of trade openness on the ability of states to combat drug trafficking. This analysis is the first attempt to assess this controversy statistically in a large sample of countries from an important region of the
world, the Americas. The empirical analysis suggests that trade ope;nness enhances the counternarcotics operations of states in drug-prpducing countries in the Americas but weakens the interdiction capabilities of states in drug-consuming countries.
Important policy implications emerge from this analysis. Should
drug-consuming countries in the Americas restrict trade in order to limit
their exposure to drug trafficking? With the growing concern over traderelated job losses and the recent collapse of the financial sector in the
United States and other OECD countries, U.S. policymakers might be
tempted to interpret this finding as additional evidence for placing
restrictions on trade. In the post-9/11 security environment, this
response may appear reasonable, especially considering that many terrorist organizations are connected to the drug trade (Drug Enforcemejnt
Administration 2002). However, it would be a mistake to erect barriers
to trade, because revenues earned from trade enable states in drug- producing countries in the Americas effectively to interdict drug trafflcking
at its very source and potentially to limit the flow of drugs into North
America. Moreover, because the analysis shows that increasing levels lof
BARTILOW AND EOM: DRUG TRAFFICKING
137
trade openness have a positive effect on the counternarcotics operations
of states in drug-producing countries in the Western Hemisphere, policymakers throughout the region should consider extending the various
trade and investment provisions of NAFTA to South America, especially
to the Andean countries, where drug trafficking is a major concern.
The results of this study also hold important implications for future
research. This research could be extended in several directions. One
possibility would be to expand the temporal domain of the analysis;
another possibility would be to expand the dataset to include the interdiction of different types of drugs from other regions of the world;
namely, Europe, Southeast Asia, Africa, and the Middle East. Such
empirical extensions of the statistical design could also be complemented with comparative case studies of the interactions between trade
openness and countries' niche in the NIDL with the ability of states to
interdict drug smuggling. As the pace of economic globalization intensifies in the years ahead, along with the growing threat of terrorism, it
will become increasingly important to understand how trade openness
affects the ability of states to interdict narcotrafficking.
APPENDIX: CALCULATION FOR DRUG ENFORCEMENT
EFFORT OF NEIGHBORING COUNTRIES
The variable Drug Enforcement Effort of Neighboring Countries was calculated with the following formula:
i-i Distance from j
* Daig Enforcement Personnel,
'
N-1
for a single country j, where N is the total number of countries, i is a
neighboring country, and i ^ j. Distance from j. is the distance from a
single country j to a neighboring country i, and drug enforcement personnel . is the number of the drug enforcement personnel in a neighboring country i. The expected relationship between the average drug
interdiction effort by neighboring countries i and a single country j is
positive. The drug enforcement personnel data were adopted from various issues of We Military Balance (Institute for Strategic Studies
1984—2006), and the distance data were collected from Direct-Line Distances, U.S edition (Eitzpatrick and Modlin 1986).
138
1J\TIN AMERICAN POLITICS AND SOCIETY
51:2
NOTES
1. Since geographic endowment enables cocaine to be produced in the
countries of the Andes, this argument is limited to the cocaine interdiction capabilities of states in the Western Hemisphere.
2. For a comprehensive discussion of how the drug trade cripples economic development and growth in source countries, see Clawson and Lee 1996.
3. Relative to the 1980s, annual cocaine consumption in the United States
declined by 50 percent in the 1990s and has since stabilized (U.N. Office bn
Drugs and Crime 2006, 96). However, during the period of this study, "in
absolute terms lU.S. cocaine consumption and] the total number of people, in
treatment for cocaine abuse tare] still by far the highest worldwide" (U.N. Office
on Drugs and Crime 2000, 59),
¡
4. Some readers may challenge these arguments on the grounds that they
need an explanation of why trade diversion matters only to transit and noti to
producing countries. NAFTA's benefits to Mexico have also been underminecl]by
the U.S. openness to the Chinese market, ,so the "diversion" factor should also
play a role in Mexico's inability to fight illegal drugs effectively. This argument
makes sense in absolute terms. Relative to drug-producing countries, however,
transit countries in Central America and the Caribbean have much sma|ler
economies of scale and are far more vulnerable to external changes in trade.
Trade diversion would certainly affect small transit countries to a greater extent
than producer countries, whose economies are far more diversified. Currently,
Mexico's oil exports have largely compensated for any diversion experienced|by
U.S, openness to the Chinese market (U.S. Congressional Budget Office 2003).
5. Cocaine interdiction data are not available for Cuba, the Cayman Islands,
and the islands of the French and Dutch West Indies.
6. To check the robustness of the results, the CIA's classification was coinipared with the U,S. State Department's International Narcotics Control Strategy Report (INCSR) classification of all the coimtries in the sample. The comparison found that there were no differences in how countries in the sample
were classified.
7. This study adopts the definition of drug enforcement personnel from the
U.S. State Department, the U.S. Drug Enforcement Administration (DEA), and
the Institute of Strategic Studies in London, which collects this information and
reports it in the publication The Military Balance. Specifically, a country's drug
enforcement personnel include its coast guard, customs agents, border police,
special drug enforcement agents attached to a country's police force, and special paramilitary drug enforcement strike forces that are attached to a country's
army, navy, and air force.
'
8. The analysis was also conducted with data on U.S. aid for economic and
military assistance, and this variable did not deviate from the results reported in
tables 1 and 2.
9. Every year the FATE produces a list of countries that are not cooperating with its recommendations for implementing anti-money-laundering measures. Most of these recommendations relate to measures that protect a coiintry's banking sector, the traditional means of laundering money. In recent years,
however, money laundering has not been limited to banks; therefore this meas-
BARTILOW AND EOM: DRUG TRAFFICKING
139
lire i.s at best conservative. To check the robustness of the results, we also used
the U.S. State Department's International Narcotics Strategy Report (1988, 1992)
classification scheme on countries whose financial systems are highly vulnerabile to money laundering. Our results are robust to the use of the INCSR
money-laundering classification scheme. These results are available from the
authors on reque.st.
10. Some readers may argue that since drug interdiction data are imperfect,
the subject cannot be studied in the context of a statistical design, and may conclude that a qualitative design would be more fruitful to study drug trafficking.
The real issue with imperfect drug interdiction data, however, is that such data
will probably produce random measurement error. But random measurement
error can be. controlled for in a large-N stati.stical design; and such error would
simply increase the uncertainty of the central explanatory variables of our study
and make the results of our research con.servative (Kmenta 1997). On the other
hand, measurement error poses a greater problem to qualitative studies of drug
trafficking because the design and data sources of these studies (archives and
historical narratives) do not allow the researcher to control for such error,
making it difficult to know if the results are merely a byproduct of uncontrolled
measurement error. In this regard, measurement error is far more likely to
undermine the integrity of qualitative studies of drug trafficking, reducing confidence that the results of such re.search are robust and rigorously derived. Qualitative studies, as a result, provide no clear advantage over large-N statistical
designs.
11. It is also possible that countries where civil liberties are high may also
have stringent criminal penalties for drug smuggling, making them less attractive for traffickers. However, we are unable to capture this phenomenon,
because C|uantitative indicators to measure the relative stringency of countries'
criminal penalties for drug smuggling are not available.
12. We also recognize that while drug traffickers can influence a country's
policymakers at the national level, their influence on policymakers at the local
and municipal levels of government may be even stronger.
13. This analysis afso included sensitivity tests in which we subjected our
model to different estimation procedures. We estimated the model using First
Order Autocorrelation with PCSE as well as Fixed Effects with PCSE. These additional estimation procedures did not alter the findings. The.se results are available from the authors on rec|uest.
14. In our sensitivity tests, the USAID variable is not significant in the fixed
effects moclel and only weakly significant in the first order autocorrelation
model.
15. One of the costs of running separate regressions fcjr different regional
subsets of the data is that some of the control variables lack sufficient variation
to generate coefficient estimates. These costs notwithstanding, the statistical estimates for trade openne.ss, the main variable of interest, are consistent with the
results reported in table 1.
140
IATIN AMERICAN POLITICS AND SOCIET\'
51:2
REFERENCES
Adam.s, Johnny [p.seudonym]. 2002. Former Drug Trafficker. Author interview.
Federal Correctional Complex, Coleman, FL, August 14.
Ahmad, Jaleel. 2004. Does Trade with Low-wage Countries Cause Wage Inequality? International Studies Review 5, 2: 81-93.
Anderson, Malcolm, ed. 1995. Policing the European Union. Oxford: Clarendon
Press.
Anderson, Thomas P. 1988. Politics and the Military in Honduras. Current History 87, 535: 425-31.
Andreas, Peter. 1998. Smuggling Wars: Law Enforcement and Law Evasion in a
Changing World. Transnational Organized Crime 4, 2: 75-90.
;
Bagley, Bruce M. 1989. The New Hundred Years War? U.S. National Security and
the War on Drugs in Latin America. In The Latin American Narcotics Trade
and tJ.S. National Security, ed. Donald J. Mabry. New York: Greenwoiod
Press. 44-58.
Batchelor, A. 2006. Lieutenant, Coast Guard Division, Jamaica Defen.se Force.
Author interview. Kingston, November 25.
;
Beck, Nathaniel. 2001. Time-Series Cros.s-Section Data: What Have We Learned
in the I^ast Few Years? Annual Review of Political Sciejice 4: 271-93.
Beck, Nathaniel, and Jonathan Katz. 1995. What to Do (and Not to Do) with TinieSeries Gross-Section Data. American Poiiticai Science Review 89: 634-47. '
Berdal, Mats, and Monica Serrano, eds. 2002. Transnational Organized Crime
and International Security: Business as Usual? Boulder: Lynne Rienner. '
Berry, William, and Stanley Feldman. 1985. Multiple Regression in Practice. Beverly Hills: Sage.
¡
Biersteker, Thomas J. 1981. Distortion or Development? : Contending Perspectives
on the Multinational Corporation. Gambridge: MIT Press.
Cabana, Mike. 2006. Director General, Border Integrity, Federal and Interriational Operations, Royal Canadian Mounted Police. Author interview.
Ottawa, June 19.
Central Intelligence Agency (CIA). Various dates. World Fact Book. See section
on transnational issues for information on country's classification in the
global drug industry. <www.theodora.com/wfb/abc_world_fact_book.
html>. Accessed October 3, 2007.
•
Clawson, Patrick L., and Renssalaer W. Lee III. 1996. The Andean Cocaine
Industry. New York: St. Martin's Press.
Clinton, William Jefferson. 1999. Speech at Conference on Progressive Governance in the 21st Century. Florence, Italy, November 21.
Colombian National Drug Council, ed. 1994. National Plans for Overcoming the
Drug Problem. Bogotá: National Drug Council.
Graig, Raymond. 2006. Senior Superintendent of Police, Organized Grime and
Firearm Bureau, Trinidad and Tobago Police Service. Author interview. Port
of Spain, February 10.
i
Graig, Richard B. 1980. Human Rights and Mexico's Anti-drug Gampaign. Social
Science Quarterly 60, 4: 691-701.
!
Grawford, Gordon, and Peter Reuter. 1988. Simulation of Adaptive Response: A
Modei of Drug Interdiction. Santa Monica, GA: RAND Gorporation.
BARTILOW AND EOM: DRUG TRAEFICKING
141
Den Boer, Monica. 2002. Law-Enforcement Cooperation and Transnational Organized Crime in Europe. In Berdal and Serrano 2002. 103-16.
Den Boer, Monica, and Neil Walker. 1993. European Policing After 1992. fournal of Common Market Studies 31, 1; 3-28.
Dillin, John. 1989. Andeans Hooked on Drug Money. Cbristian Science Monitor, March 7: 7.
Dodwell, David, and Canute James. 1993. GATT Attacks Lomé Convention Over
Bananas. Financial Times, June 16: 1-5.
Drake, Paul W. 1998. The International Causes of Democratization. In The Origins of Liberty: Political and Economic Liberalization in the Modern World,
ed. Drake and Mathew D. McCubbins. Princeton: Princeton University
Press. 70-91.
Dreher, Axel. 2006. Does Clobalization Affect Growth? Evidence from a New
Index of Globalization. Applied Economics iQ, 10: 1091-110.
Edwards, L. 2006. Major, Coast Guard Division, Jamaica Defense Force. Author
interview. Kingston, November 24.
Eeige, Edgar L. 1996. Overseas Holdings of U.S. Currency and the Underground
Economy. In Pozo 1996. 5-59.
Financial Action Task Force on Money Laundering (FATF). 2002. Report on
Money Laundering Typologies, 2001-2002.
Paris: FATF Secretariat,
OECD.
. 2003. Report on Money Laundering Typologies, 2002-2003. Paris: FATF
Secretariat, OECD.
Fitzpatrick, Gary L., and Marilyn J. Modlin. 1986. Direct-Line Distances. U.S. ed.
Metuchen: Scarecrow Press.
Flynn, Stephen E. 1993. Wodd Wide Drug Scourge: The Expanding Trade in
Illicit Drugs. Brookings Review 11, 1: 6-11.
Freedom House. 2007. Ereedom in tbe World Country Ratings, 1972-2007.
<www.freedomhouse.org>. Accessed September 23, 2007.
Frey, Bruno S., and Werner W. Pommerehne. 1982. Measuring the Hidden Economy: Though This Be Madness, There Is Method in It. In Tanzi 1982b. 3-27.
Galtung, Fredrick. 2000. A Global Network to Curb Corruption: The Experience
of Transparency International. In Tbe Third Eorce: The Rise of Transnational Civil Society, ed. Ann M. Florini. Washington, DC: Carnegie Endowment for International Peace. 17-47.
Garcia, Jorge Ipseud.]. 2002. Director, Anti-Narcotics, Colombian National Police.
Author interview. Bogota, November 20.
George, Susan. 1992. 'Tbe Debt Boomerang: How Third World Debt Harms Us All.
Boulder: Westview l^ress.
Gill, Stephen. 1996. Globalization, Democratization, and the Politics of Indifference. In Globalization: Critical Reflections, ed. James Mittleman. Boulder:
Lynne Rienner. 205-88.
Green, Jeff Ipseud.]. 2002. Eormer Drug Trafficker. Author interview. Federal
Correctional Complex, Coleman, FL, August 14.
Held, David, and Anthony McGrew, eds. 2000. 'The Global Transformations
Reader: An Introduction to tbe Globalization Debate. Maiden: Polity Press.
Helson, Gary D. 2004. Supervisory Special Agent, U.S. Drug Enforcement
Admini.stration. Author interview. San Diego, July 14.
142
LATIN AMFRICAN POUTICS AND SOCIETi'
51:2
Hinds, A. C. P. 2006. Assistant Commi.ssioner of Police, Jamaica Constabulary
Force. Author interview. Kingston, November 21.
Hufbauer, Gary, and Jeffrey). Schott. 1992. North A/nerica Free Trade: Issues and
Recommendations Washington, DC: Institute for International Economics.
HLirreii, Andrew, and Ngaire Woods, eds. 1999. Inequaiity, Giobaiization and
Worid Pointes. New York: Oxford Univensity Press.
Hutchinson, Giad.stone, and Ute Schmacher. 1996. NAFTA's Threat to Central
American and Caribbean Ba.sin Exports: A Revealed Comparative Advantage Approacii. fournai of Interamerican Studies and Worid Affairs 36, 1
(Spring): 127^8.
Internationai Institute for Strategic Studies. 1984-2006. The Miiitary Balance.
Variou.s issues. London: Internationai Iastitute for Strategic Studie.s.
Jaggers, Keith, and Ted Rollert Gurr. 1993. Tracking Democracy's Third Wave
with the Polity III DÍÜH. fournai of Peace Research 32, 4: 409-82.
Jamaica Daily Gleaner (Kingston). 1996. Caribbean Drugs Problem Could
Worsen. May 14: 6.
Johnson, Harry G. 1970. The Efficiency and Welfare Implications of the International Corporation. In The Internationai Corporation: A Symposium, ëd.
Charles P. Kindleberger. Cambridge: MIT Press. 35-56.
Kaufmann, Daniel, and Shang-Jin Wei. 1999. Does "Grease Money" Speed Up
the Wheels of Commerce? Working Paper 7093. Cambridge, MA: National
Bureau of Economic Research.
¡
Kerry, John. 1998. The New War: The Web of Crime that Threatens America's
Security. New York: Touchstone.
'
Kmenta, Jan. 1997. Eiements of Econometrics. 2nd ed. Ann Arbor: University of
Michigan Press.
;
Lawrence, Robert. 1996. Single World, Divided Nations: Internationai Trade and
OECD Labor Markets. Washington, DC: Brookings Institution Pre.ss.
Lee, Ren.sselaer W. III. 1989. The White Labyritith: Cocaine and Poiiticai Power.
New Brunswick: Transaction.
Leff, Nathaniel H. 1964. Economic Development Through Bureaucratic Coraiption. American Behaviorai Scientist 8, 2: 8-14.
Lepgold, Joseph. 1998. Hypotheses on Vulnerability: Are Terrori.st and Drug
Traffickers Coercible? In Strategic Coercion: Concepts and Cases, e'd.
Lawrence Freedman. New York: Oxford University Press. 131-50.
Levi, Michael. 2002. Liberalization and Transnational Financial Crime. In Berdal
and Serrano 2002. 53-66.
'
Lewin, Hardley. 2006. Rear Admiral, Coa.st Guard Division, Jamaica Defense
Force. Author interview. Kingston, November 24.
Li, Quan, and Rafael Reuveny. 2003. Economic Globalization and Democracy:
An Empirical Analysis. British Journal of Political Science 55, 1: 29-54. '
Lui, Francis. 1983. An Equilibrium Queuing Model of Bribery. Journai of Poiiticai Economy 93, 4: 760-81.
Mabiy, Donald J., and Raphael Ped. 1989. Concluding Observations and Policy
Recommendations. In The Latin American Narcotics Trade and U:S.
Nationai Security, ed. Mabry. New York: Greenwood Press. 151-61.
!
Marshall, Monty G., and Keith Jaggers. 2002. Poiity IV Project: Dataset tJsei^'
Manual. College Park, MD: Center for International Development and Conflict Management.
'
BARTILOW AND EOM: DRUG TRAEEICKING
143
Martin, Philip L, 1993. Trade and Migration: NAFTA and Agriculture. Policy
Analyses in International Economics 38. Washington, DC: Institute for International Economics.
McLaughlin, Eugene. 1992. The Democratic Deficit: European Union and the
Accountability of the Bdtish Police. British Journal of Criminology 32, 4:
473-87.
Mittleman, James. 1996. The Dynamics of Globalization. In Globalization: Critical Reflections, ed. Mittleman. Boulder: Lynne Rienner. 1-19.
Molefsky, Barry. 1982. America's Underground Economy, In Tanzi 1982b. Al-(fl.
Monshipouri, Mahmood. 1995. Democratization, Liberalization, and Human
Rights in the Third World. Boulder: Lynne Rienner.
Naylor, R. T. 2002. Wages of Crime: Black Markets, Illegal Einance, and the
Underworld Economy. Ithaca: Cornell University Press.
Ogden, Derek, 2006. Chief Superintendent, Director-General of Drugs and Organized Crime, Federal and International Operations, Royal Canadian
Mounted Police, Author interview. Ottawa, June 19.
Organization for Economic Cooperation and Development (OECD) Statistics
Directorate; International Monetary Fund; Bureau of Statistics; International
Labor Organization; Bureau of Statistics, Commonwealth of Independent
States. Statistical Committee. 2002. Measuring the Non-Observed Economy:
A Handbook. Paris: OECD.
Organization of American States (OAS). Inter-American Drug Abu,se Control
Commission (CICAD). n.d. Data on annual cocaine interdiction in Western
Hemisphere countries. <www.cicad.oas.org/oicl/Estadisticas/default,htm>
Accessed September 19, 2007.
Paul, Trevor. 2006. Commissioner of Police, Trinidad and Tobago Police Service.
Author interview. Port of Spain, February 9.
Perl, Raphael F. 1993-94. Clinton's Foreign Drug Policy. Journal of Interamerican Studies and World Affairs 35, 4 (Winter): 143-51.
Polity IV Project. 2007. Political Regime Characteristics and Transitions, 1800-2007.
<www.systemicpeace.org/polity/polity4.htm> Accessed October 25, 2007,
Pozo, Susan, ed. 1996. Exploring the Underground Economy: Studies of Illegal
and Unreported Activity. Kalamazoo: W, E. Upjohn Institute for Employment Research.
Rosenberg, Matthew J. 2000. Jamaica Battling Cocaine Smuggling. Mountain
Eagie (Whitesburg, Kentucky), February 16: 6.
Rother, Larry. 1997. Backlash From NAFTA Batters Economies of Caribbean.
Neiv York Times, January 30.
Rueschemeyer, Dietrich, and Peter Evans. 1985. The State and Economic Transformations: Towards an Analysis of the Conditions Underlying Effective
Intei"vention. In Bringing the State Back In, ed, Rueschmeyer, Evans, and
Theda Skocpol. Cambridge: Cambridge University Press. 69-92,
Serrano, Monica. 2002. Transnational Organized Crime and International Security: Business as Usual? In Berdal and Serrano 2002. 13-35.
Stares, Paul. 1996. Global Habit: The Drug Problem in a Borderless World. Washington, DC: Brookings Institution Press.
Tanzi, Vito. 1982a. Underground Economy and I'ax Evasion in the United Stales:
Estimates and Implications. In Tanzi 1982b. 69-92.
144
Ij\TIN AMERICAN POLITICS AND SOCIE'lT
51: 2
Tanzi, Vito, ed. 1982b. The Underground Economy in the United States and
Abroad. Toronto: Lexington Books.
Taylor, Ian. 2002. Liberal Markets and the Republic of Europe: Contextualizing
the Growth of Transnational Organized Crime. In Berdal and Serrano 2002.
119-53.
i
Teeple, Gary. 1995. Globalization and the Decline of Social Reform. Atlantic
Highlands, NJ: Humanities Press.
Tokatlian, Juan G. 1994. The Miami Summit and Drugs: A Placid, Innocuous
Conference? Journal of Interamerican Studies and World Affairs 36, 3 (Fall):
75-91.
'
;
Transparency International Perception Corruption Index. 1995-2008. <www.
transparency.org/policy research/sui-vey.s_indices/cpi> Accessed September
1,2007.
Treverton, F. Gregory. 1999. International Organized Grime, National Security
and the Market State, tn Transnational Crime in the Americas, ed. Tom
Farer. New York: Routledge. 39-56.
i
Tulder, Frank van. 1992. Crime, Detection Rate, and the Police: A Macro
Approach, fournal of Quantitative Criminology 8, 1: 113-31.
United Nations Office on Drugs and Grime. 2000. World Drug Report 2000.
United Nations.
;
. 2006. World Drug Report 2006. United Nations.
U.S. Agency for International Development. U.S. Overseas Loans and Grants,
1946-2006. Data on U.S. aid given to Western Hemisphere countries to support drug interdiction operations, 1984-2003. <http://qesdb.cdie.org/gbk/
index.html> Accessed August 15, 2007.
U.S. Gongre.ssional Budget Office. 2003. The Effects of NAFTA on U.S.-Mexican
Trade and G13P. Report. Washington, DG: Gongressional Budget Office.
<www.cbo.gov/ftpdocs/42xx/doc4247/Repon.pdf
U.S. Gustoms and Border Protection. 2000a. The Commissioner's Message. tJ.S.
Customs Today, July: 1-2.
i
. 2000b. The Gonstant Ghange in Goncealment Methods. U.S. Custorns
Today, July: 3-4.
U.S. tDepartment of Defense. 1991. FDH General Staff Officer Reponecily
Involved in Narcotic Trafficking. Declassified Memorandum. Washington,
DG: U.S. Department of Defen.se.
U.S. Department of State, n.d. Data on U.S. maritime drug interdiction agfe'ements with Western Hemisphere countries. <www.state.gov/s/1/2005/
87199.htm> Accessed March 15, 2009.
. 1988. International Narcotics Control Strategy Report. Washington, DC:
State Department.
'•
. 1992. International Narcotics Gontrol Strategy Report. Washington, DG:
State Department.
U.S. Drug Enforcement Administration (DEA), n.d. Data on U.S. maritime drug
interdiction agreements with We.stern Hemisphere countries. <www.usdój.
gov/dea/pubs/intel/01019/usvi#usvi>
I
. 2002. Statement by Asa Hutchinson, Administrator, Drug Enforcemeiit
Administration, Before the Senate Judiciary Gommittee Subcommittee on
Technology, Terrorism, and Government Information. March 13. <www.
u.sdoj.gov/dea/pub.s/cngrte.st/ct031302p.html> Accessed March 13, 2007. ;
BARTILOW AND EOM: DRUG TRAFFICKING
145
U.S. Drug Enforcement Administration (DEA). Strategic South American/
Caribbean Unit. 2001. The Drug Trade in the Garibbean: A Threat Assessment. August. DEA document no. 01019. Washington, DG: DEA. <www.
usdoj.gov/dea/pubs/intel/01019/01019.html> Accessed January 10, 2007.
Vernon, Raymond. 1975. Multinational Enterprises in Developing Countries: An
Analysis of National Policies. New York: United Nations Industrial Development Organization.
Wallerstein, Immanuel. 1974. The Modern World System. New York: Academic
Press.
. 1984. The Politics of the World Economy. Gambridge: Gambridge University Press.
Walters, Joe [pseud.]. 2001. Financial Investigator, U.S. Drug Enforcement
Administration. Author interview. New York, June 11.
Watkis, Devon. 2006. Senior Superintendent, Organized Grime Branch, Jamaica
Gonstabulary Force. Author interview. Kingston, November 21.
Williams, Phil. 1994a. Transnational Griminal Organizations and International
Security. Survival 36, 1: 96-113.
. 1994b. Transnational Organized Grime: Strategic Alliances. Washington
Quarterly 18, 1: 57-72.
. 2002. Gooperation Among Griminal Organizations. In Berdal and Serrano
2002. 67-82.
Wisotsky, Steven. 1986. Breaking the Impasse in the War on Drugs. Westport:
Greenwood Press.
World Bank. 2006. Worid Development Indicators, 2006. Washington, DG: World
Bank.